Download Clarifying Adverse Drug Events

Improving Patient Care
Clarifying Adverse Drug Events: A Clinician’s Guide to Terminology,
Documentation, and Reporting
Jonathan R. Nebeker, MS, MD; Paul Barach, MD, MPH; and Matthew H. Samore, MD
Adverse drug events cause substantial morbidity and mortality, yet
they remain underappreciated and misunderstood. The terminology to describe errors and patient harm associated with medications causes much confusion. This article uses the case study of a
patient with multiple adverse drug events to clarify key terms,
such as adverse event, adverse drug reaction, adverse drug event,
medication error, and side effect. The case discussion illustrates
clinical approaches to analyzing the causal connection between a
suspect drug and an adverse event. Examples and rationale for
meaningful documentation of adverse drug events are provided,
along with an outline of the types of events that should be
reported to regulatory agencies.
S
ibuprofen. Is this event a side effect, an adverse drug reaction, a medication error, or an exacerbation of his underlying renal and cardiac disease?
Terms that initially arose from the field of pharmacovigilance, such as adverse event and adverse drug reaction, can help physicians relate the edema and renal
failure to ibuprofen. Pharmacovigilance is the study of
drug-related injuries for the purpose of making warning
or withdrawal recommendations for pharmaceutical
products. The International Conference on Harmonisation of Technical Requirements for Registration of
Pharmaceuticals for Human Use, of which the U.S.
Food and Drug Administration (FDA) and the World
Health Organization are members, defines an adverse
event as “any untoward medical occurrence that may
present during treatment with a pharmaceutical product
but which does not necessarily have a causal relationship
with this treatment” (15). The term adverse event is not
particularly helpful to physicians, but it provides context for the more clinically useful term adverse drug reaction. The International Conference on Harmonisation
defines an adverse drug reaction as “a response to a drug
which is noxious and unintended and which occurs at
doses normally used in man for prophylaxis, diagnosis,
or therapy of disease or for the modification of physiologic function” (15). Therefore, an adverse drug reaction is an adverse event with a causal link to a drug.
Table 1 (16 –20) summarizes these key terms.
ince the early 1990s, adverse drug events have received
significant attention from researchers in quality and patient safety (1). Nationally recognized quality experts have
identified adverse drug events as a top safety priority (2)
because these events are the most common type of iatrogenic injury (3). Studies have indicated that adverse drug
events occur almost daily in medium-sized hospitals and
outpatient panels (4 – 6). However, despite the high morbidity and mortality, physicians often do not recognize or
appropriately treat instances of drug-related harm (7, 8).
We believe that inadequate recognition and treatment
of drug-related harm are, in part, a result of what has been
called a Tower of Babel of terminology (1). Terms originally developed in the narrow context of drug effects in a
clinical and regulatory setting are now being applied in the
broader context of quality improvement in health care delivery systems (9). As might be expected, the expanding
role of these terms has been coupled with their use in
contradictory ways, even within the same discipline (4, 7,
10 –14). In this paper, we use the case of an actual patient
as a framework to explain the recognition, treatment, documentation, and reporting of drug-related harm.
ADVERSE EVENTS
VERSUS
ADVERSE DRUG REACTIONS
Mr. J. was a 70-year-old man with nephrotic syndrome
(thought to be related to a congenital single kidney), pneumoconiosis, and a history of gout and myocardial infarction. He
presented to the hospital with increasing bilateral leg edema
and pain, for which he had been taking over-the-counter ibuprofen, 400 mg three times a day for 3 days and once a day for
the preceding 3 weeks. His other outpatient medications were
simvastatin, 40 mg at bedtime; aspirin, 81 mg once daily; and
metoprolol, 50 mg twice daily. In the emergency department,
his serum creatinine level was 680 ␮mol/L (7.7 mg/dL),
much higher than the baseline of 290 ␮mol/L (3.3 mg/dL) 11
months earlier. He was admitted to the hospital.
The patient experienced an adverse event while using
Ann Intern Med. 2004;140:795-801.
For author affiliations, see end of text.
www.annals.org
ASSESSING CAUSAL ASSOCIATIONS
After admission to the hospital, a battery of serologic tests,
a microscopic urine examination, and abdominal ultrasonography did not yield specific information about the origin of
Mr. J.’s exacerbated renal failure.
Assessing causal connections between agents and disease is fundamental to the practice of medicine and to the
understanding of adverse drug reactions (21, 22). In this
Improving Patient Care is a special section within Annals supported in part by the U.S. Department of Health and Human Services (HHS) Agency for Healthcare Research and Quality
(AHRQ). The opinions expressed in this article are those of the authors and do not represent the position or endorsement of AHRQ or HHS.
© 2004 American College of Physicians 795
Improving Patient Care
Clarifying Adverse Drug Events
Table 1. Summary of Definitions Relevant to Drug-Related Harm
Term
Definition*
Example
Harm occurred
Adverse event
Harm in a patient administered a drug but not necessarily caused by a drug (16)
Traumatic death while taking
lovastatin
Congestive heart failure from
metoprolol
Adverse drug reaction
Adverse drug event
Harm may have occurred
Medication error
Side effect
Harm did not occur
Potential adverse drug event
Harm directly caused by a drug at normal doses† (16)
Unexpected adverse drug reaction:
An adverse drug event whose nature or severity is not consistent with the
product information (17)
Harm caused by the use of a drug (4, 18)
Effective definition in common practice:
Harm caused by a drug or the inappropriate use of a drug
Inappropriate use of a drug that may or may not result in harm (19)
Hematoma from tirofiban
overdose
A usually predictable or dose-dependent effect of a drug that is not the principal
effect for which the drug was chosen; the side effect may be desirable, undesirable,
or inconsequential (17)
Failure to renew prednisone order
on transfer to medical ward
(This term should be avoided
when considering adverse
events)
Circumstances that could result in harm by the use of a drug but did not harm the
patient
Receipt of roommate’s felodipine
but no resulting hypotension
* Definitions are abstracted from cited sources. See text for original definitions.
† The International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use recently dropped the dose limits for
adverse drug reactions, but it is not clear whether or when the U.S. Food and Drug Administration will adopt these revised definitions (20).
Table 2. Grades of Certainty That an Event Is Linked to
a Drug*
Level
Criteria
Certain
A clinical event, including an abnormal laboratory test
result, that occurs in a plausible time relationship to
drug administration and cannot be explained by
concurrent disease or other drugs or chemicals. The
response to withdrawal of the drug (dechallenge)†
should be clinically plausible. The event must be
pharmacologically or phenomenologically definitive,
with use of a satisfactory rechallenge† procedure if
necessary.
Probable/Likely
A clinical event, including an abnormal laboratory test
result, that occurs within a reasonable time
sequence to administration of the drug, is unlikely
to be attributed to concurrent disease or other
drugs or chemicals, and follows a clinically reasonable response on withdrawal (dechallenge)†. Rechallenge† information is not required to fulfill this definition.
Possible
A clinical event, including abnormal laboratory test
result, that occurs within a reasonable time
sequence to administration of the drug but could
also be explained by concurrent disease or other
drugs or chemicals. Information on drug withdrawal
may be lacking or unclear.‡
Unlikely
A clinical event, including an abnormal laboratory test
result, whose temporal relationship to drug administration makes a causal relationship improbable and
in which other drugs or chemicals or underlying
disease provides plausible explanations.
* Adapted with permission from Elsevier (The Lancet, 2000;356:1255-9) (16).
† For adverse drug events caused by withdrawal for or reduced dose of a drug,
dechallenge is restoring the previous drug dose and rechallenge is reducing the drug
dose or withdrawing the drug again.
‡ Although an adverse drug event may rate only as “possible” soon after discovery,
it can be flagged as requiring more information and later be upgraded to probable
or certain as appropriate.
796 18 May 2004 Annals of Internal Medicine Volume 140 • Number 10
case, the treatment and prognosis for Mr. J.’s renal failure
largely depend on whether it was exacerbated by ibuprofen
or was due to the progression of his underlying disease.
The discipline of pharmacovigilance has yielded tools
(16, 23–25) to assess the likelihood of a causal connection
between a drug and an adverse event on a case-by-case
basis. These tools address the following criteria: time relationships between the drug use and the adverse event,
pathophysiology of the adverse event, competing causes for
the adverse event, response to dechallenge (for example,
discontinuation of therapy with the drug or dose reduction), and response to rechallenge (for example, drug readministration). Table 2 organizes these criteria to gauge
the causal link between a drug and an adverse event in
terms of 4 discrete levels of certainty (certain, probable/
likely, possible, and unlikely).
The causality criteria listed in Table 2 can be applied
to the case of Mr. J. The timing and classic pathophysiologic association of ibuprofen with the edema and worsening renal failure may first seem to result in a certain causal
association. However, other explanations for the event, such
as advancing intrinsic renal failure or cardiac disease, are
present, and information on the effect of drug withdrawal
is not available at this point in the hospitalization. These
conditions result in a possible causality rating, which accurately captures the uncertainty in the causal analysis. The
strength of a causal association may be revised as more
information becomes available.
The day after admission, Mr. J. developed painful, swollen
joints. After sodium urate crystals were found in the synovial fluid,
polyarticular gout was diagnosed and prednisone therapy was
started. Despite administration of a 1-L normal saline challenge
and subsequent high-dose furosemide, the patient remained oliguwww.annals.org
Clarifying Adverse Drug Events
ric with an elevated creatinine level. Before dialysis could begin,
he became hypertensive; the metoprolol dose was increased to 100
mg twice a day. Within 2 hours of the increased metoprolol dose,
the patient developed respiratory distress and was transferred to
the medical intensive care unit, where he was intubated and
pulmonary edema was diagnosed.
Admitted because of 2 adverse events— edema and renal failure—Mr. J. developed 4 subsequent adverse events:
a gout flair, hypertension, pulmonary edema, and respiratory distress. A consideration of pathophysiologic pathways
and causal associations can also help the physician identify
adverse drug reactions in complex clinical scenarios.
The basic unit of an adverse event is the sequence of
pathophysiologically related events originating from one pharmacologic effect of the drug. In Mr. J.’s case, the presumed
pharmacologic effect was the inhibited production of prostaglandin in the afferent glomerular arteriole, which led to decreased plasma filtration. The consequent events of uric acid
retention, sodium retention, and fluid retention all resulted
from ibuprofen’s inhibition of prostaglandin production.
In general, a drug may be considered a contributory cause
of an adverse event if, had the drug not been administered, 1)
the event would not have happened at all, 2) the event would
have occurred later than it actually did, or 3) the event would
have been less severe. Ibuprofen may have contributed to the
edema and may have exacerbated renal failure. The fluid bolus
increased the fluid overload and may have caused hypertension. The subsequent increase in metoprolol probably contributed to cardiac decompensation and pulmonary edema. In
contrast, the gout flair is a renal failure–mediated secondary or
indirect effect (17) of ibuprofen and does not merit consideration as a separate adverse drug reaction.
MEDICATION ERROR
After spending 2 days in the medical intensive care unit
and having several liters of fluid removed by dialysis, Mr. J.
was extubated and prepared for transfer to the medical ward.
Just before transfer, he became tachycardic and diaphoretic; an
electrocardiogram showed new precordial T-wave inversions.
After consultation with the cardiologist, the intern prescribed
12.5 ␮g of tirofiban. The pharmacy prepared and the nurse
administered 12.5 mg—a thousand-fold overdose. The patient developed a 3-cm hematoma on the back of his hand at
a previous venipuncture site and oozing from his dialysis catheter site. He underwent urgent, prolonged dialysis and multiple blood laboratory checks.
Misreading physician orders and preparing a dangerous dose of a drug is a classic medication error. Medication
error is commonly defined as any preventable event that
may cause or lead to inappropriate medication use or patient harm while the medication is in the control of the
health care professional, patient, or consumer. Such events
may be related to professional practice, health care products, procedures and systems, including prescribing; order
communications; product labeling, packaging and nomenwww.annals.org
Improving Patient Care
clature; compounding; dispensing; distribution; administration; education; monitoring; and use. (19)
Most medication errors do not harm patients. Some
authors estimate that less than 1% of medication errors result
in harm (26). Some types of errors may present minimal potential for harm, such as a missed dose of simvastatin. Other
types of errors pose substantial risk to the patient but are
intercepted before reaching the patient; these have been called
near misses, close calls, or potential adverse drug events (27). For
example, the nurse who was about to administer the tirofiban
to Mr. J. might have noticed that the dose was inappropriate
and could have returned the drug to the pharmacy.
ADVERSE DRUG EVENT
REACTION
VERSUS
ADVERSE DRUG
Several days after Mr. J. was transferred back to the medical ward, his gout recurred. It was discovered that the intern
did not include the prednisone prescription when she wrote
orders for the patient’s transfer back to the medical ward.
The overdose of tirofiban and the recurrence of gout
are 2 medication errors that help illustrate the differences
between an adverse drug reaction and an adverse drug event.
An adverse drug reaction occurs at usual doses and is caused
by the action of the drug, such as renal failure due to ibuprofen. Because the hematoma from tirofiban was the result of an
abnormally high dose, it does not qualify as an adverse drug
reaction. The gout recurrence was a consequence of the unmasking of the underlying disease rather than the action of a
drug itself, so this too is not an adverse drug reaction. Pharmacovigilance, as practiced by regulatory bodies, is primarily
concerned with adverse drug reactions—the properties of the
drug under normal use. The patient safety community is interested in harm resulting from a broader range of events, such
as these 2 medication errors. The broader interests of the patient safety movement have reinforced the need for a term
other than adverse drug reaction and have led to the adoption
of the term adverse drug event.
The use of the term adverse drug event is consistent
both between the pharmacovigilance and patient safety
communities and within the patient safety community itself. The FDA recognizes the term adverse drug event to be
a synonym for adverse event, in which a causal association
may not exist between the event and the drug (28). In the
patient safety literature, the terms adverse drug event and
adverse event usually denote a causal association between
the drug and the event (3), but there is a wide spectrum of
definitions for these terms, including harm caused by a
drug (7, 10), harm caused by drug use (4), and a medication error with or without harm (29).
We recommend the definition of adverse drug event
adopted by the Institute of Medicine: “an injury resulting
from medical intervention related to a drug” (3, 4), which
has been simplified to “an injury resulting from the use of
a drug” (18). Under this definition, the term adverse drug
event includes harm caused by the drug (adverse drug re18 May 2004 Annals of Internal Medicine Volume 140 • Number 10 797
Improving Patient Care
Clarifying Adverse Drug Events
Figure. Relationships of key terms.
The gray areas represent injuries caused by drug use (adverse drug
events). The dark gray area represents harm caused by a drug (adverse
drug reactions). The light gray area represents harm from appropriate
drug use that is generally excluded from studies of adverse drug events.
Medication errors are much more common than adverse drug events, but
they result in harm less than 1% of the time (30). Conversely, about one
quarter of adverse drug events are due to medication errors (4).
actions and overdoses) and harm from the use of the drug
(including dose reductions and discontinuations of drug
therapy). The term adverse drug event does not include
failure to use a drug in the first place, which is not a use of
a drug. Although this definition is broad, the patient safety
literature generally limits adverse drug event by excluding
nonserious injuries resulting from appropriate dose titration and disease recurrences resulting from appropriate reductions or discontinuations of therapy with the drug.
These implicit limitations according to appropriateness
have resulted in a close relationship between the terms
medication error and adverse drug event (medium gray area
in the Figure) (30).
RECOGNIZING
AND
TREATING ADVERSE DRUG EVENTS
The day after the tirofiban overdose, the patient reported
lower abdominal pain. Since admission, the patient had been
taking narcotics and, when not intubated, calcium carbonate,
1.25 g 3 times a day. Nursing flow sheets and dietary notes
indicated almost daily that, despite eating, the patient had not
passed stool since admission. On hospital day 13, the patient
was given 1 dose of milk of magnesia and began receiving
docusate sodium. The patient responded with 1 small bowel
movement. Indomethacin was started, presumably to reduce
the patient’s narcotic requirement. Several days later, Mr. J.
vomited and was given droperidol. Otherwise, Mr. J. was
improving and was discharged from the inpatient medicine
service to the rehabilitation service. The next morning the
patient had a second bowel movement—5 days after the first.
About an hour later, the patient vomited during breakfast,
developed respiratory distress, and was reintubated.
Physicians fail to recognize a majority of adverse drug
events (7, 8) for many reasons. Many physicians find nursing notes difficult to read and may ignore these data, but
798 18 May 2004 Annals of Internal Medicine Volume 140 • Number 10
nursing data may provide the only indication for 40% of
all adverse drug events (31). Physicians also commonly
classify gastrointestinal adverse drug reactions as “side effects” and believe them to be common and unavoidable
consequences of medical care, not clinically significant
manifestations of disease. Constipation from narcotics and
nausea from antibiotics are examples of adverse drug reactions that frequently occur but seldom cause serious outcomes. However, even though only a small percentage of
these events are serious, these “side effects” are so common
that serious manifestations, as exemplified by Mr. J.’s case,
are not rare (4, 7, 18, 32). The international pharmacovigilance community has recognized that the term side effect tends to minimize the injury from drugs and has recommended that this term no longer be used (17).
The failure to recognize an adverse drug reaction as
such may lead the physician to inappropriately treat the
adverse drug event (33). In this case, the physician did not
pursue a diagnosis for the nausea and abdominal pain but
instead treated the symptoms with droperidol. The pervasive problem of treating each successive drug-related symptom with another medication exposes the patient to additional drug hazards. In this case, the patient was put at risk
for sedation, delirium, or cardiac dysrhythmias from the
droperidol. To avoid multiple adverse drug events, the
more appropriate action often is to discontinue therapy
with the original drug.
DOCUMENTING ADVERSE DRUG EVENTS
The purpose of documenting adverse drug events in a
patient’s chart is to help prevent the recurrence of the
harm. The more likely an event is to recur and the more
serious the event, the stronger the case for documentation
and the stronger the incentive to guard against similar use
of the drug in the future. Events that are likely to recur are
those that have a high causal association with the drug and
that result from a common use of the drug. Thus, probable
and certain adverse drug reactions are nearly always appropriate to document in the patient chart. For events that are
less likely to recur, the physician must consider the seriousness and nature of the event. For example, a life-threatening, possible adverse drug reaction merits documentation.
However, documentation is unlikely to prevent the recurrence of an adverse drug event due to an unusual error,
such as the tirofiban overdose. These types of errors are
better reported through an institution’s incident reporting
system.
Documentation is most likely to prevent future adverse drug events when it serves either to caution against
the use of a drug at any dose or to establish practical,
unsafe dosing ranges. Allergies are common examples of
idiosyncratic reactions that make the drug’s use inappropriate at any dose. Dose-dependent reactions are nearly 8
times more frequent than idiosyncratic reactions (7); not
including dose information in documentation is a major,
www.annals.org
Clarifying Adverse Drug Events
missed opportunity for injury prevention. In Mr. J.’s example, metoprolol was well tolerated at 25 and 50 mg
twice a day but not at higher doses. (A second episode of
pulmonary edema also occurred after he received two
50-mg doses several hours apart.) Although high doses are
more likely to cause adverse drug reactions (4), doses that
are too low might also unmask disease. When drugs are
suppressing a chronic disease, such as a dysrhythmia or
seizure disorder, and the disease recurs after discontinuation of therapy with the drug or dose reduction, the unsafe
dose should be documented to avoid unmasking the disease again in the future.
A meaningful and useful record of an adverse drug event
combines all 3 of these factors: the severity of the event, the
causal association between the drug use and the event, and
dosing variables. In this case, for example, intravenous morphine at 2 mg 3 to 4 times daily possibly led to severe constipation (no stool for 13 days) and probably contributed to
nausea and vomiting that resulted in aspiration and intubation. Contributors were calcium carbonate for the constipation and indomethacin for nausea and vomiting.
Finally, it is important to document adverse drug
events in an accessible part of the medical record. Although
many physicians discuss the differential diagnosis of adverse drug events in their progress notes, they may note
only the drug’s name in the allergy section of the chart. A
more detailed description in the allergy section may help
another physician decide whether subsequent use of the
drug is contraindicated. Because most hospitalizations and
clinic visits do not involve an adverse drug event, the burden of improved documentation is small. Table 3 provides
other examples of meaningful documentation for some of
Mr. J.’s adverse drug reactions. In all, Mr. J. had at least 10
adverse drug events; none of his events were documented
in the allergy or adverse drug event section of the medical
record.
Mr. J.’s respiratory status rapidly improved, and he was
extubated the next day. Shortly afterward he had a dark,
guaiac-positive stool; therapy with indomethacin was stopped.
Two days later and hours before anticipated discharge, the
patient vomited and aspirated his breakfast. He developed
extensive bilateral pneumonitis and pneumonia. Mr. J. died
the next evening.
REPORTING
TO
REGULATORY AGENCIES
It might appear that at least one of Mr. J.’s adverse
drug events should be reported to the FDA, but this is not
the case. Documenting adverse drug events in the patient’s
chart and reporting adverse drug reactions to regulatory
agencies are distinctly different activities (Table 4). The
FDA is interested in receiving reports on serious, unexpected adverse drug reactions (not adverse drug events)
from marketed drugs. Unexpected reactions are those
whose nature or severity is not consistent with the product
label (17). The FDA and international regulatory bodies
www.annals.org
Improving Patient Care
Table 3. Suggested Components and Examples of Meaningful
Documentation of Adverse Drug Events
Components
Drug, dose (for dose-dependent adverse drug events), causality assessment, description of events/outcomes, including contributing factors
Examples
Ibuprofen, 400 mg 3 times daily, possibly exacerbated renal failure,
with creatinine level increasing from 290 to 680 ␮mol/L (3.3 to 7.7
mg/dL) and 3⫹ edema.
Metoprolol, 100 mg twice daily, probably precipitated congestive heart
failure and respiratory failure within 2 hours of the increased dose:
once in the setting of renal failure, fluid overload, and aspiration and
once in the setting of aspiration alone. Metoprolol, 25–50 mg twice
daily, was well tolerated.
Intravenous morphine, 2 mg 3–4 times daily, possibly led to severe
constipation and probably contributed to nausea and vomiting resulting in aspiration and intubation. Contributors were calcium carbonate
for constipation and indomethacin for nausea and vomiting.
Indomethacin, 50 mg 3 times daily, possibly led to gastritis, nausea,
and vomiting, resulting in aspiration and intubation. Prednisone may
have contributed to gastritis, and morphine may have contributed to
nausea and vomiting.
(The hematoma from the tirofiban overdose and the gout recurrence
from the inadvertently discontinued prednisone are medication errors
that should not be documented in the adverse drug event section of
the patient’s medical record.)
define serious events as those resulting in death, life-threatening experiences, prolonged or initial hospitalization, significant or persistent disability, or a congenital anomaly or
requiring intervention to prevent one of these outcomes
(28). The FDA has discouraged reports on events that are
neither unexpected nor serious, except for events associated
with newly marketed drugs (34). While nearly all of Mr.
J.’s adverse drug events were serious, none is convincingly
unexpected in nature or severity.
The FDA has set up a voluntary reporting system for
physicians called MedWatch that encourages and facilitates
the reporting of serious, unexpected adverse drug reactions.
MedWatch forms can be obtained at www.fda.gov
/medwatch/report/hcp.htm or through the pharmacy departments at most hospitals. Online reporting is also available at the preceding Web site.
After a drug is marketed, physicians may be the only
source of information the FDA has for rare and potentially
fatal adverse drug events. However, physicians frequently neglect this crucial role. As few as 1% of serious and unexpected
events are estimated to be reported to the FDA (35). Underreporting can lead to substantial delays in dissemination of
warnings and product labeling changes (36). Most institutions facilitate reporting by making pharmacists available to complete and submit these reports. If a pharmacist is not available in the outpatient setting, the
physician can use online tools to fill out a MedWatch
report; personnel at the FDA will communicate with the
physician if the report needs clarification or correction.
18 May 2004 Annals of Internal Medicine Volume 140 • Number 10 799
Improving Patient Care
Clarifying Adverse Drug Events
Table 4. Comparison of Documentation and Reporting of Adverse Drug Events*
Events and Reports
Document in Patient’s Chart
Types of events
Drug-related injuries
Probable or certain adverse drug reactions
Report to FDA
Life-threatening, possible adverse drug events
Dosing ranges specific to the patient that resulted in
adverse drug events
Error/potential adverse drug
events
Properties of reports
Goal
Determination of causation
Where to report
Voluntary or mandatory
Adverse drug events resulting from medication errors
that may be repeated with normal drug use
Not relevant
Prevent recurrence of an adverse drug event in one
patient
Physician determines casual link between drug and
event for each case
Physician documents in allergy or adverse drug event
section of patient chart
Required for good clinical care
Unexpected and serious adverse drug reactions:
Serious reactions result in death, life-threatening experiences, prolonged or initial hospitalization, clinically
significant or persistent disability, or a congenital
anomaly or require intervention to prevent one of the
above (28).
Not usually relevant (Confusing drug labeling is an
exception.)
Contribute to labeling or withdrawal recommendations
FDA determines causation from multiple cases. Physician
provides information to facilitate this determination.
Physician (or, for inpatients, usually pharmacist) submits
MedWatch report to FDA: www.fda.gov/medwatch
Voluntary but encouraged
* FDA ⫽ U.S. Food and Drug Administration.
CONCLUSION
AND
TAKE-HOME POINTS
We propose that a better understanding of terms will
help clinicians recognize and treat drug-related injuries
(Table 5). Adverse drug reactions are injuries caused by
drugs administered at usual doses; they are the primary
focus of regulatory agencies and postmarketing surveillance. Adverse drug events are injuries caused by drug use
that encompass adverse drug reactions and harm resulting
Table 5. Key Points for Definitions, Causal Assessment,
Documentation, and Reporting
Definitions
Adverse event and adverse drug reaction are regulatory terms; the first
does not require a causal link between the drug and the event, the
second does.
Adverse drug events extend beyond adverse drug reactions to include
harm from overdoses and underdoses usually related to medication
errors. A minority of adverse drug events are medication errors, and
medication errors rarely result in adverse drug events.
The term side effect should be avoided.
Causal assessment: grades of causation
Certain: dechallenge and rechallenge information corroborates causation
Probable: dechallenge information corroborates causation
Possible: competing explanations are plausible
Unlikely: timeline is improbable
Documentation
Documentation helps determine whether subsequent use of a drug in
possibly different circumstances is contraindicated.
from medication errors; they are the targets of broader
efforts to improve patient safety.
All adverse drug events, including those that are common symptoms, warrant careful attention and evaluation,
rather than rote management via standing orders. Nursing
documentation is a rich source of information about adverse drug events that manifest as patient discomfort.
The clinical value of accurate documentation of adverse drug events is that it assists future determinations of
whether the risks of prescribing a specific drug or drug class
outweigh the drug’s potential benefits for an individual
patient. Providing information on the causal association of
the drug and the event is therefore useful. Physicians or
pharmacists are strongly encouraged to report unexpected,
serious adverse drug reactions to the FDA using the MedWatch system, which requires information similar to but
more extensive than that recommended for documentation
in a patient’s chart (37).
Health care institutions must integrate allergy and adverse drug event reports into the care delivery process. The
first step is to expand allergy documentation to accommodate all types of adverse drug events. An important next
goal is to integrate the reports electronically into the orderchecking process to give nursing and pharmacy personnel
the opportunity to intercept potential adverse drug events
before they occur.
Considerations of seriousness and causation should drive the decision
about whether to document adverse drug reactions.
Meaningful descriptions of adverse drug events include the grade of
causation, a description of the event or outcome, contributing or
competing factors, and dose information.
Reporting
Serious and unexpected adverse drug events should be reported to the
U.S. Food and Drug Administration.
Information for causal assessment is a crucial and often neglected portion of reports.
800 18 May 2004 Annals of Internal Medicine Volume 140 • Number 10
From Veterans Affairs Salt Lake City Health Care System and University
of Utah School of Medicine, Salt Lake City, Utah; and Jackson Memorial Hospital, University of Miami, Miami, Florida.
Grant Support: By the Geriatric Education and Clinical Center
(GRECC) of the Veterans Affairs Salt Lake City Health Care System and
by Health Services Research & Development grants SAF 98-122, TRP
02-147, and RCD 02-176.
www.annals.org
Clarifying Adverse Drug Events
Acknowledgments: The authors thank Adam R. Nebeker, MD, and
Seth R. Lewis, MD, for their critical reviews of the manuscript.
Potential Financial Conflicts of Interest: None disclosed.
Requests for Single Reprints: Jonathan R. Nebeker, MS, MD, Salt
Lake Informatics, Decision Enhancement And Surveillance (IDEAS)
Center, GRECC 182, 500 Foothill Drive, Salt Lake City, UT 84148;
e-mail, [email protected].
Current author addresses are available at www.annals.org.
References
1. Ross SD. Drug-related adverse events: a readers’ guide to assessing literature
reviews and meta-analyses. Arch Intern Med. 2001;161:1041-6. [PMID: 11322836]
2. Shojania KG, Duncan BW, McDonald KM, Wachter RM. Safe but sound:
patient safety meets evidence-based medicine [Editorial]. JAMA. 2002;288:50813. [PMID: 12132985]
3. Kohn LT, Corrigan JM, Donaldson MS. To Err Is Human: Building a Safer
Health System. Washington, DC: National Academy Pr; 1999.
4. Bates DW, Cullen DJ, Laird N, Petersen LA, Small SD, Servi D, et al.
Incidence of adverse drug events and potential adverse drug events. Implications
for prevention. ADE Prevention Study Group. JAMA. 1995;274:29-34. [PMID:
7791255]
5. Gandhi TK, Weingart SN, Borus J, Seger AC, Peterson J, Burdick E, et al.
Adverse drug events in ambulatory care. N Engl J Med. 2003;348:1556-64.
[PMID: 12700376]
6. Fattinger K, Roos M, Vergeres P, Holenstein C, Kind B, Masche U, et al.
Epidemiology of drug exposure and adverse drug reactions in two swiss departments of internal medicine. Br J Clin Pharmacol. 2000;49:158-67. [PMID:
10671911]
7. Classen DC, Pestotnik SL, Evans RS, Burke JP. Computerized surveillance of
adverse drug events in hospital patients. JAMA. 1991;266:2847-51. [PMID:
1942452]
8. Tegeder I, Levy M, Muth-Selbach U, Oelkers R, Neumann F, Dormann H,
et al. Retrospective analysis of the frequency and recognition of adverse drug
reactions by means of automatically recorded laboratory signals. Br J Clin Pharmacol. 1999;47:557-64. [PMID: 10336581]
9. Nebeker JR, Hurdle JF, Hoffman JM, Roth B, Weir CR, Samore MH.
Developing a taxonomy for research in adverse drug events: potholes and signposts. J Am Med Inform Assoc. 2002;9(6 Suppl):S80-5.
10. Schlienger RG, Luscher TF, Schoenenberger RA, Haefeli WE. Academic
detailing improves identification and reporting of adverse drug events. Pharm
World Sci. 1999;21:110-5. [PMID: 10427579]
11. Johnstone DM, Kirking DM, Vinson BE. Comparison of adverse drug
reactions detected by pharmacy and medical records departments. Am J Health
Syst Pharm. 1995;52:297-301. [PMID: 7749958]
12. Sivaram CA, Johnson S, Tirmizi SN, Robertson V, Garcia D, Sorrells E.
Morning report: a forum for reporting adverse drug reactions. Jt Comm J Qual
Improv. 1996;22:259-63. [PMID: 8743062]
13. Thatcher C. ADR: reporting problems. Can Pharm J. 2002;134:11.
14. Hofer TP, Kerr EA, Hayward RA. What is an error? Eff Clin Pract. 2000;
3:261-9. [PMID: 11151522]
15. Clinical Safety Data Management: Definitions and Standards for Expedited
Reporting. London: European Agency for the Evaluation of Medicinal Products,
Human Medicines Evaluation Unit; 1995. Accessed at www.emea.eu.int/pdfs
/human/ich/037795en.pdf on 16 December 2003.
16. Edwards IR, Aronson JK. Adverse drug reactions: definitions, diagnosis, and
management. Lancet. 2000;356:1255-9. [PMID: 11072960]
www.annals.org
Improving Patient Care
17. Cobert BL, Biron P. Pharmacovigilance from A to Z. Malden, MA: Blackwell Science; 2002.
18. Gurwitz JH, Field TS, Avorn J, McCormick D, Jain S, Eckler M, et al.
Incidence and preventability of adverse drug events in nursing homes. Am J Med.
2000;109:87-94. [PMID: 10967148]
19. What Is a Medication Error? Rockville, MD: National Coordinating Council
for Medication Error Reporting and Prevention; 1998. Accessed at www.nccmerp
.org on 29 August 2003.
20. Post-Approval Safety Data Management: Definitions and Standards for Expedited Reporting E2D. Geneva, Switzerland: International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for
Human Use; 2003. Accessed at www.ich.org/MediaServer.jser?@_ID⫽631&
@_TYPE⫽MULTIMEDIA&@_TEMPLATE⫽616&@_MODE⫽GLB on 17
December 2003.
21. Venulet J, ten Ham M. Methods for monitoring and documenting adverse
drug reactions. Int J Clin Pharmacol Ther. 1996;34:112-29. [PMID: 8705089]
22. Meyboom RH, Hekster YA, Egberts AC, Gribnau FW, Edwards IR. Causal
or casual? The role of causality assessment in pharmacovigilance. Drug Saf. 1997;
17:374-89. [PMID: 9429837]
23. Michel DJ, Knodel LC. Comparison of three algorithms used to evaluate
adverse drug reactions. Am J Hosp Pharm. 1986;43:1709-14. [PMID: 3752106]
24. Kramer MS, Leventhal JM, Hutchinson TA, Feinstein AR. An algorithm
for the operational assessment of adverse drug reactions. I. Background, description, and instructions for use. JAMA. 1979;242:623-32. [PMID: 449002]
25. Naranjo CA, Busto U, Sellers EM, Sandor P, Ruiz I, Roberts EA, et al. A
method for estimating the probability of adverse drug reactions. Clin Pharmacol
Ther. 1981;30:239-45. [PMID: 7249508]
26. Bates DW. Medication errors. How common are they and what can be done
to prevent them? Drug Saf. 1996;15:303-10. [PMID: 8941492]
27. Schaaf T, Van der Lucas DA, Hale AR, eds. Near Miss Reporting as a Safety
Tool. Oxford: Butterworth-Heinemann; 1991.
28. Guideline for Postmarketing Reporting of Adverse Drug Experiences. Rockville, MD: U.S. Food and Drug Administration; 1992. Accessed at www.fda.gov
/medwatch/safety/ade/t_cder.htm on 16 December 2003.
29. Andrews LB, Stocking C, Krizek T, Gottlieb L, Krizek C, Vargish T, et al.
An alternative strategy for studying adverse events in medical care. Lancet. 1997;
349:309-13. [PMID: 9024373]
30. Bates DW, Boyle DL, Vander Vliet MB, Schneider J, Leape L. Relationship
between medication errors and adverse drug events. J Gen Intern Med. 1995;10:
199-205. [PMID: 7790981]
31. Hurdle JF, Weir CR, Roth B, Hoffman J, Nebeker JR. Critical gaps in the
world’s largest electronic medical record: ad hoc nursing narratives and invisible
adverse drug events. Proc AMIA Symp. 2003:309-12. [PMID: 14728184]
32. Hanlon JT, Schmader KE, Koronkowski MJ, Weinberger M, Landsman
PB, Samsa GP, et al. Adverse drug events in high risk older outpatients. J Am
Geriatr Soc. 1997;45:945-8. [PMID: 9256846]
33. Evans RS, Pestotnik SL, Classen DC, Bass SB, Burke JP. Prevention of
adverse drug events through computerized surveillance. Proc Annu Symp Comput Appl Med Care. 1992:437-41. [PMID: 1482913]
34. Lucas LM, Colley CA. Recognizing and reporting adverse drug reactions.
West J Med. 1992;156:172-5. [PMID: 1536067]
35. Wood AJ, Stein CM, Woosley R. Making medicines safer—the need for an
independent drug safety board. N Engl J Med. 1998;339:1851-4. [PMID:
9854125]
36. Ladewski LA, Belknap SM, Nebeker JR, Sartor O, Lyons EA, Kuzel TC, et
al. Dissemination of information on potentially fatal adverse drug reactions for
cancer drugs from 2000 to 2002: first results from the research on adverse drug
events and reports project. J Clin Oncol. 2003;21:3859-66. [PMID: 14551305]
37. Ahmad SR. Adverse drug event monitoring at the Food and Drug Administration. J Gen Intern Med. 2003;18:57-60. [PMID: 12534765]
18 May 2004 Annals of Internal Medicine Volume 140 • Number 10 801
Current Author Addresses: Dr. Nebeker: Salt Lake Informatics, Decision Enhancement And Surveillance (IDEAS) Center, GRECC 182, 500
Foothill Drive, Salt Lake City, UT 84148.
Dr. Barach: Department of Anesthesiology, Jackson Memorial Hospital,
University of Miami, North Wing 109, 1611 NW 12th Avenue, Miami,
FL 33136.
E-802 © American College of Physicians
Dr. Samore: Division of Clinical Epidemiology, Salt Lake Informatics,
Decision Enhancement And Surveillance (IDEAS) Center, 50 North
Medical Drive, Salt Lake City, UT 84132.